A survey to build framework for optimize and secure migration and transmission of cloud data

ABSTRACT


INTRODUCTION
The emergence of cloud technologies comes with advantages and disadvantages.Cloud technology is beneficial in various factors over dedicated servers like security, load balancing (LB), and optimization, while some are challenges, such as the number of customers using the cloud.IT industries are developing and rigorously utilizing the cloud for application development and data storage, which come up with different questions related to the same factors as security and all.Real-time applications need fast computation and data access; overburdening secure layers may lose purpose [1], [2].
The recent decade's development in cloud computing has created various research in these areas, which needs different architectures and frameworks in cloud technology, which may enable modeling, simulation, and infrastructure to be secure, sustainable, and optimized for various applications and their needs.Most cloud users need fast, secure, easy customization, and configuration without considering technological backend features such as migration, LB, and virtual machine (VM) management.Some recent development uses intelligence in different layers to provide needed features like heuristic in migration to nodes.Other uses swarm intelligence for it.These intelligent algorithms help in providing fast LB in VMs.Cloud services (infrastructure as a service (IaaS), platform as a service (PaaS), and software as a service (SaaS)) have other challenges and need different solutions.Cloud service providers (CSP) must consider all these issues and challenges [3], [4].

ISSN: 2302-9285 
A survey to build framework for optimize and secure migration … (Ravinder Bathini)

813
Hence, we are motivated to provide a generic framework to provide state-of-the-art solutions for different layers of problems.The exhaustive survey indicates different parameters to provide various types of solution integration in the framework.This integration with the optimal utility to maintain fast responses is a key feature of the proposed framework.The application-specific customization is another advantage of the framework, so need-based customization and optimization is a standout feature of the proposed framework.

PROPOSED METHOD
A proposed framework for optimal end-to-end secure access shown in Figure 1.From the literature survey, as per need, the optimal best algorithm selection module is provided in the framework.Existing end-to-end transmission control protocol/internet protocol (TCP-IP) protocol-based security is identified.Optimal secure algorithms are incorporated to provide layer-based end-to-end security.The secure protocol suite with two-way data storage encryption will improve security on different brute force and middleman attacks.The LB algorithm selection provides an optimal algorithm selection per resource utilization and VM's ideal time.
Similarly, recovery algorithm selection is based on deployment, user, and cloud type.This will improve access and security to data and applications.The significant improvement with this framework is algorithm selection for fast response and maintaining end-to-end security.
Figure 1.End-to-end optimal cloud security framework

OPEN CLOUD
As discussed in the introduction, several cloud services are classified into IaaS, PaaS, and SaaS-based on their utilities and the cloud services shown in Figure 2. Here, the literature survey mainly focuses on fault tolerance and recovery, LB, and security.Some open-source cloud simulators are being used for experimental purposes.Their features are described in Table 1.Based on their utilities, different services in the ever-changing cloud computing landscape are divided into three categories: SaaS, PaaS, and IaaS.These divisions are essential in defining the features and services provided by cloud services, as Figure 2 shows.This survey's investigation of fault tolerance and recovery techniques in cloud systems is one of its main goals.Given the inherent complexity of distributed systems, it is critical to comprehend how cloud services manage errors and bounce back from disturbances.The assessment assesses best practices and current methodology in this field, illuminating novel strategies to improve system resilience.Understanding and developing cloud services require experimentation.In light of this, the survey investigates the use of free and open-source cloud simulators for testing.The features of these simulators are listed in Table 1, which also provide an overview of their capabilities and how they support the creation and testing of cloud-based solutions.

Table 1. Features of cloud simulators
Simulator Features Language Type GUI support Stimulation time CloudSim [4], [5] Java Open source Limited Second Green cloud [6] C++ Open source Limited (support via nam) Minute iCanCloud [7] C++ Open source Not limited Second

LOAD BALANCING
It is one of the key tasks in cloud computing, sometimes called migration.Applications and data are being migrated from nodes or VMs based on policies mentioned in Figure 3. Location, information, selection, and transfer-based policies represent different load-balancing needs.Location policy indicates geographical data migration, like country data can be migrated from one to another for fast access.Similarly, other policies are designed for the migration of data [8], [9].

Figure 3. LB policies
There are LB techniques shown in Figure 4. Based on system state and initiation, there are sub-types discussed.Sender, receiver, and symmetric initiation typically represent LB initiation started.The same algorithmic strategies used for static and dynamic LB are shown under system state-based LB [10]- [12].Tables 2 and 3 shows LB algorithm comparison.4. It takes into account things like scalability, consistency of data, and recovery time.It's critical for firms to comprehend these attributes in order to select a CSP that best suits their unique recovery needs.Fault tolerance policies play a crucial role in guaranteeing the robustness of cloud infrastructures, even beyond data recovery.Table 5 lists the various fault tolerance strategies that various architectures have chosen.This covers methods to deal with system failures and DR tactics.Organizations that want to continue operating smoothly even in the face of unforeseen difficulties must evaluate these policies.The ability of a company to recover from data loss and survive system outages is greatly impacted by the cloud service provider that it chooses.Organizations can make decisions that are relevant to their needs by comparing the recovery and fault tolerance rules shown in Tables 4 and 5.
6. SECURITY Security is the most crucial issue in dedicated server vs cloud comparison.Different aspects need to be considered before comparison.There are some characteristics [50] of cloud storage given in Table 6.Based on these storage characteristics, data is stored in the cloud.The stored data needed to be protected; hence, there are some security groups and problems [51] described in Table 7. Table 8  There is a growing controversy in the field of data management regarding dedicated servers vs cloud storage.Security is a key component of this conversation since businesses want to safeguard their important data.Prior to comparing cloud computing to dedicated servers, it is critical to take into account a number of factors that influence the security paradigm.The fundamental features of cloud storage are listed in Table 6, which forms the basis for cloud data management.These attributes-which range from accessibility to scalability-have a significant impact on the storage market.Still, security continues to be the main focus, highlighting the necessity of strong defenses.Cloud data storage necessitates a closer attention to security measures.Table 7 explores security groups and possible issues pertaining to cloud storage.Every component, including access controls and encryption techniques, helps to protect the confidentiality and integrity of data that is stored.

CONCLUSION
The proposed framework provides layer-based end-to-end protection for data and applications in the cloud.A selection of load-balancing algorithms at the runtime provides optimal migration.These dynamic algorithm selection modules for LB, fault tolerance framework and recovery help optimally execution for migration, security, and data transmission.The various certification as per domain security ensures data privacy.A significant improvement will be observed in algorithm selection as per the need of resources and system, VM ideal time.This will improve the throughput of the system without compromising security.The proposed framework is an open security structure that includes two-way protection at each layer of end-to-end security.The performance major is still a significant concern and will be calculated based on different parameters for domains like security, migration, and throughput.


ISSN: 2302-9285 Bulletin of Electr Eng & Inf, Vol. 13, No. 2, April 2024: 812-820 816 CSPs use several recovery techniques to lessen the effects of data loss.A thorough comparison of these systems based on their salient characteristics is shown in Table

Table 2 .
LB procedures comparison A survey to build framework for optimize and secure migration … (Ravinder Bathini) 815

Table 3 .
LB algorithms comparison continue Different CSPs use different recovery mechanisms in case of data loss.A comparative analysis of such mechanisms is given in Table4based on their properties.Similarly, different architectures have different policies for fault tolerance compared to Table5for disaster recovery (DR).

Table 4 .
Cloud recovery mechanism and properties

Table 5 .
Cloud fault tolerance comparison

Table 6 .
represents the cloud security framework comparison.Cloud storage characteristics

Table 8 .
Cloud security framework comparison